THE
d-AMINO
ACID
OXIDASE
OF NEUROSPORA
BY N. H. HOROWITZ
(From the School of Biological
Sciences, Stanford
California)
March 9, 1944)
Among artificially produced mutants of the mold Neurospora have been
found strains lacking the ability to synthesize specific amino acids (1, 2).
In the course of biochemical and genetic studies of this group of mutants it
was observed that some of the mutants, e.g. those deficient in methionine,
leucine, and arginine,’ are able to utilize racemic mixtures of the amino acids
with the same efficiency as the 1, or physiologically occurring, forms. In
the casesof the leucine- and the methionine-requiring mutants it was also
possible to show utilization of the ar-keto analogues. It thus appeared possible that the mode of conversion of the d to the 1 isomers consists in oxidative deammation, followed by resynthesis. A study was therefore undertaken to test the ability of Neurospora to oxidize the “unnatural” optical
isomers of the amino acids. It was found that extracts of the mold contain
a d-amino acid oxidase similar in its action to the d-amino acid oxidase of
mammalian kidney and liver (3). This finding supports the above hypothesis for the conversion of the d- to the l-amino acids.
Since it appears that the d-amino acid oxidase has not been previously
described in fungi, a number of experiments were performed on the Neurospora enzyme, the results of which are reported here.
Methods
Wild type Neurospora crassawas grown in Fernbach flasks containing 500
ml. of the salt-sucrose-biotin medium previously described (4). After 7 to
14 days at 25” the pads were harvested and washed in several changes of the
basal salt medium. They were then pressedout through a cloth to remove
excesswater and weighed. At this stage the pads weighed 4 to 6 gm. each
and contained 70 to 75 per cent of water. They were next ground in a
mortar, with sand and 2 ml. of ~/60 pyrophosphate buffer, pH 8.5, per gm.
of wet tissue. The resulting paste was centrifuged at high speed for several
minutes, and the supernatant, containing the enzyme, was poured off and
diluted with 0.25 volume of 0.25 M pyrophosphate, pH 8.5. The final
pH, determined with the glass electrode, was 8.0 to 8.2.
Oxygen consumption was measured in the Warburg apparatus at 28.6”.
1 See the papers on the leucineless mutant (Regnery, D. C., J. Biol. Chem., 164,
151 (1944)) and on the arginineless mutants (Srb, A. M., a.nd Horowitz,
N. H., J.
Biol. Chem., 164, 129 (1944)).
141
Downloaded from www.jbc.org at CALIFORNIA INSTITUTE OF TECHNOLOGY on December 26, 2006
(Received for publication,
University,
142
$-AMINO
ACID
OXIDASE
OF
NEUROSPORA
Results
Stoichiometric Relations-In
the absence of added substrates the oxygen
consumption of the preparation is slight but measurable.
On the addition
of dl-methionine a rapid oxidation was observed. The rate of oxygen con-
FIG.
Curve
valine;
cal for
FIG.
(Curve
FIG. 1
FIG. 2
1. Oxidation
of some amino acids by Neurospora d-amino acid oxidase.
A, dl-methionine;
Curve B, dl-leucine; Curve C, dl-isoleucine;
Curve D, dlCurve E, dl-lysine; Curve F, dl-ornithine.
The horizontal line is the theoretithe uptake of 1 atom of oxygen per molecule of one optical isomer.
2. pH curves of hreurospora enzyme on dl-methionine
(Curve A), dl-alanine
B), and dl-a-amino-n-caprylic
acid (Curve C).
sumption remained almost constant until 0.25 mole of oxygen per mole of
dl-methionine was taken up, and then it rapidly dropped to zero. When
I-methionine was substituted for the racemic mixture, no oxidation occurred.
It is thus evident that the reaction involves the oxidation of d-methionine
only, with the uptake of 1 atom of oxygen per molecule (Fig. 1). The same
relation was found to hold for all other d&amino acids whose oxidation
rate was high enough to make an accurate determination
of the end-point
readily possible; namely, dl-phenylalanine,
dl-norvaline,
dl-citrulline,
dl-arginine, dl-ac-amino-n-butyric
acid, dl-leucine, dl-norleucine, dl-isoleutine, and dl-glutamic acid.
The keto acid analogue of methionine, cY-keto-y-methiolbutyric
acid, was
found to be a product of the oxidation of d-methionine by the enzyme. It
Downloaded from www.jbc.org at CALIFORNIA INSTITUTE OF TECHNOLOGY on December 26, 2006
2 ml. of the enzyme solution were placed in the main compartment and 0.2
ml. of a ~/15 solution of the racemic amino acid in the side arm. In the
case of insoluble amino acids, a solution of the sodium salt was used. KOH
was placed in the well; the atmosphere was air.
In all experiments the small autorespiration was automatically
corrected
for by placing enzyme solution in the thermobarometer
vessel.
N.
H.
HOROWITZ
143
Downloaded from www.jbc.org at CALIFORNIA INSTITUTE OF TECHNOLOGY on December 26, 2006
was isolated from the reaction mixture in the form of its 2,4-dinitrophenylhydrazone, melting at 149”, in agreement with the melting point published
by Waelsch and Borek (5) and by Cahill and Rudolph (6). When dissolved
in alkali the compound gave the red color characteristic
of the 2,4-dinitrophenylhydrazones
of a-keto.acids.
Sulfur (by sodium fusion) was present,
and sulfhydryl
(by the nitroprusside
test) was absent.
pH Optimum-The
effect of pH changes in the range pH 6 to 10 on the
activity of the enzyme was determined.
Phosphate buffer was used at pH
6 to 8, pyrophosphate
at pH 9 to 10. Determinations
were made on three
different substrates,
dl-methionine,
dl-alanine, and dl-ar-amino-n-caprylic
acid, respectively.
In all cases a marked optimum at pH 8.0 to 8.5 was
observed (Fig. 2).
Effect of Substrate Concentration-The
relation between substrate concentration and reaction rate, with dl-methionine
as substrate, was found to
follow the usual hyperbolic law, within experimental limits.
The Michaelis
constant was approximately
2.5 X 10-*. This value represents the concentration of d-methionine which produces the half maximum velocity, and is
equal to the dissociation constant of the enzyme-substrate
complex.
Inhibitors-The
system is not significantly inhibited by cyanide (0.001
M), iodoacetate (0.001 M), or benzoate (0.01 M). Benzoate has been reported to produce complete inhibition of the kidney d-amino acid oxidase
at a concentration of 0.01 in (7). On the other hand, drying the tissue with
acetone and ether before extracting does not affect the activity of the mammalian enzyme, but in the case of Neurospora this treatment results in
inactive preparations.
The Neurospora enzyme is competitively
inhibited
by isovaline (see below).
SpeciJicity-The
enzyme was found to oxidize the d forms of most of the
amino acids tested.
Glycine and l-amino acids, with the exception of l-glutamate, are not oxidized.
Z-Glutamate is oxidized at less than one-fifth the
rate of d-glutamate under the conditions of these experiments and presumably by a different enzyme system.
As is the case with the d-amino acid oxidase of kidney, d-methionine is the
substrate most readily attacked by the Neurospora enzyme.
The oxygen
uptake on dl-methionine
(6.06 X low3 M) of sixteen different preparations
varied from 64.2 to 148 c.mm. of oxygen per hour per gm. of wet weight of
mold, with a mean value of 107 c.mm. The cause of the variability is not
definitely known.
The experiments
have indicated, however,
that the
variat,ion in activity does not affect the relative rates of oxidation of the
amino acids. In the determination
of the oxidation rates preseuted in
Table I the activity of each new enzyme preparation was standardized
on
dl-methionine
as substrate,
to which all other substrates
were then referred.
144
&AMINO
ACID
OXIDASE
0~
~EUR0sp0R.4
TABLE
Relative
Rates
of Oxidation
of Amino
Acids
I
by d-Amino
Acid
Oxidase
of Neurospora
The mean rate of oxidation of dl-methionine
= 107 c.mm. of 02 per hour per gm.
of wet mold. All amino acids were tested in a final concentration
of 3.03 X lo-* M
in terms of one optical isomer.
-
-
GA%tive
rate
Substrate
dl-Methionine.
............
100 dl-N-Methylleucine.
.................
dl-Phenylalanine
...........
85 dl-c+Aminophenylacetic
acid .........
dl-Norvaline.
..............
85 dl-Tryptophane.
....................
dl-Citrulline.
..............
81 dl-Ornithine
.........................
dl-Arginine ................
80 dl-Serine ............................
dl-cu-Amino-n-butyric
acid .
74 dl-Threonine........................
dl-Leucine. ................
66 dl-Proline ...........................
dl-Norleucine.
.............
52 @Alanine ...........................
dl-Glutamic
acid ...........
41 dl-fl-Amino-n-butyric
acid ...........
dl-Isoleucine. ..............
38 dl-oc-Amino-oc-methylbutyric
acid .....
d( -)-Alanine
.............
33 dl-ru-Amino-a-ethylbutyric
acid ......
dl-Aspartic acid. ..........
29 dl-fi,&Dimethyl-a-amino-n-butyric
acid ..............................
dl-Alanine ..................
26
dl-Saline. .................
26 dZ-N,N-Dimethylleucine
.............
dl-cu-Amino-n-caprylic
acid . 22 dl-Leucylglycine
....................
dl-Lysine. .................
14 Glycine .............................
-
lelativerate
Substrate
13
About 9
“ 5
“ 4
0
0
0
0
0
0
0
-
lowering the reactivity of the substrate has also been noted in studies of
the mammalian
d-amino acid oxidase (8-10).
Inhibition by Isovaline-A number of the non-reactive amino acids were
tested for their effect on the oxidation of methionine.
If these substances
attach to the enzyme to form an inactive complex, they should competitively inhibit the oxidation of other amino acids. If, on the other hand,
no or only slight complex formation occurs, no inhibition is expected. The
dl-/3following compounds were tested : dl-serine, d&N, N-dimethylleucine,
amino-n-butyric
acid, and dl-isovaline
(a-amino-a-methylbutyric
acid).
No inhibition of methionine oxidation was found with the first three, even
at concentrations which were 10 times higher than the concentration
of
Downloaded from www.jbc.org at CALIFORNIA INSTITUTE OF TECHNOLOGY on December 26, 2006
As can be seen from Table I, the following changes in the structure of the
substrates destroy their reactivity: shift of the amino group from the CYto
the /3 position; replacement of the hydrogen attached to the a-carbon atom
by an alkyl group; replacement by methyl groups of both hydrogens
attached to the amino nitrogen atom; replacement by methyl groups of
the hydrogens attached to the o-carbon atom; substitution of a hydroxyl
group on the P-carbon atom; and peptide bond formation through the
carboxyl group. The effect of substitutions
on the p-carbon atom in
N.
H.
145
HOROWITZ
” = K,Kr
VW) Ki
+ K,(Z) + K&9
(1)
TABLE
II
Inhibition
of Neurospora Enzyme by Zsovaline
The concentrations
of amino acids are given in terms of one optical isomer. The
isovaline concentration
was 3.0 X 10e2 M in all experiments.
A fresh preparation
of
enzyme was used for each experiment.
Experiment
No.
ILkthionine
8
concentration
Ki
.-
M x 103
x 108
3.0
3.0
1.5
1.5
0.75
3.0
1.5
0.75
0.75
Mean
0.36
0.36
0.52
0.44
0.46
0.36
0.53
0.69
0.69
. . .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1
4.1
4.0
5.4
8.8
4.1
3.8
3.4
3.4
4.6
where v’ = the rate of inhibited reaction, V = the maximum rate (proportional to the enzyme concentration), Ki = the dissociation constant of the
enzyme-inhibitor complex, K, = the dissociation constant of the enzymesubstrate complex, (8) = the substrate concentration, and (1) = the
inhibitor concentration. In the absence of inhibitor the rate is given by
the Michaelis-Menten equation,
(2)
Combining the above equations, one obtains for the inhibited fraction of
the rate, p,
I’
p=v!=
V
K
KsKi
(1)
+ KS(Z) + K~@‘)
Downloaded from www.jbc.org at CALIFORNIA INSTITUTE OF TECHNOLOGY on December 26, 2006
methionine.
It is concluded that in these cases complex formation with
the enzyme does not occur.
In the case of the fourth substance tested, isovaline, an inhibition of
methionine
oxidation
was observed.
The competitive
nature of the
inhibition is indicated by its dependence on the concentration
of methionine (Table II).
The dissociation constant of the enzyme-isovaline
complex was calculated by a modification of the equation of Lineweaver
and
Burk (ll),
146
d-AMINO
ACID
OXIDASE
OF
NEUROSPORA
from which
Ki = Km (1 - P)
PUG + (6’))
(3)
R-CHNHZ-COOH
+ +O, --j R-C-COOH
II
NH
+ Hz0 +
R-CO-COOH
+ NHa
TABLE
III
Effect of Chain Length on Reactivity of Straight Chain Amino Acids toward Neurospora
Enzyme
Final concentration
of amino acids, 3.03 X 10e3 M in terms of one optical isomer.
Atmosphere,
air; temperature,
28.6”.
Oxyge;5c~;;med
Substrate
in
c. mm.
dl-Alanine.....................................................
dl-ol-Amino-n-butyricacid.....................................
dZ-Norvaline...................................................
dZ-Norleucine..................................................
dk-Amino-n-caprylic
acid.. . . . . . . . . . . . . . . . . . . . .
..
9.4
17.2
21.4
16.9
6.2
In the cases of serine, N,N-dimethylleucine,
and /3-amino-n-butyric acid,
imino formation, or its equivalent,2 is theoretically possible, but the reaction is blocked by factors which prevent attachment of the molecule to the
enzyme. In the mammalian d-amino acid oxidase, Keilin and Hartree (12)
have shown that neither a-methylalanine nor N , N-dimethylalanine is
able to form a complex with the enzyme.
.E#ect of Chain Length-An important relation shown in Table I concerns the effect of chain length on the reactivity of substrates toward
Neurospora enzyme. With increasing length of the carbon chain in the
2 The corresponding
quaternary ammonium
oxidation product
salt, R-C(-COOH.
CH,--N-CH3
+
of N,N-dimethylleucine
would
be the
Downloaded from www.jbc.org at CALIFORNIA INSTITUTE OF TECHNOLOGY on December 26, 2006
Table II shows values of Ki calculated by means of Equation 3, with K, =
2.5 X 1O-4 (see above). The constancy of Ki may be considered good in
view of the errors involved in the determination of K, and of p at low
concentrations of substrate.
The failure of isovaline to be oxidized by the enzyme is ascribable to the
impossibility of forming the imino structure,
N.
H.
HOROWITZ
147
DISCUSSION
The function of d-amino acid oxidase in the metabolism of Neurospora
is unknown.
Any explanation which is based on the hypothesis that the
organism may encounter racemic amino acids in nature, or that it may
produce them in the course of the digestion and assimilation of proteins,
appears unacceptable, since the wild type of Neurosporu is able to synthesize all of its amino acids from carbohydrates and inorganic nitrogen; it
is consequently independent of external supplies of amino acids. If the
enzyme serves a useful purpose, it would therefore seem to be concerned
with products of the organism’s own metabolism.
This suggests the
possibility of symmetric synthesis of amino acids by the mold.
The applicability to Neurospora of the recent finding by Shemin and Rittenberg
(13) that d-glutamic acid and d-tyrosine are not synthesized by the riboflavin-deficient rat is an open question.
In amino acid-deficient mutants of Neurospora, present evidence suggests that the d-amino acid oxidase plays an essential part in the transformation of d-amino acids (supplied from the outside in racemic mixtures)
to Z-amino acids. Thus, d-methionine,
d-leucine, and d-arginine are all
rapidly oxidized by the enzyme and are efficiently utilized by the corresponding mutant strains. In the cases of methionine
and leucine the
evidence is more complete, in that utilization of the a-keto analogues has
also been found. The a-keto analogue of arginine has not been tested.
Further evidence, of an indirect kind, comes from the tryptophane-requiring mutants.
Tatum and Bonner (14) have shown that tryptophane synthesis in Neurospora occurs by a condensation of indole with I-serine. dl-
Downloaded from www.jbc.org at CALIFORNIA INSTITUTE OF TECHNOLOGY on December 26, 2006
homologous series of straight chain, monoaminomonocarboxylic
acids,
the oxidation rate first rises to a maximum at a length of 5 carbon atoms
(norvaline) and then drops off. Since the data in Table I were obtained
at different times, with a fresh enzyme preparation each time, it appeared
desirable to check this relation on a single preparation.
This was done
with the results shown in Table III.
These data corroborate the previous
result.
It seems clear from these findings that an optimum chain length exists
among the substrates of the Neurospora enzyme. The effect of various
substitutions and internal rearrangements on the reactivity of the substrate
may thus in part be ascribed to the changes they produce in the length of
the molecule.
Published reports do not indicate a similar dependence in
the case of crude mammalian
d-amino acid oxidase. In the case of the
purified mammalian enzyme, it appears that rate data are not available for
a sufficient number of substrates to decide the point.
1.48
d-AMINO
ACID
OXIDASE
OF
NEUROSPORA
This work was supported by grants from the Rockefeller
Foundation.
The author is indebted to Dr. David Bonner for samples of the following
amino acids: dl-proline, dl-N-methylleucine,
dl-N ,N-dimethylleucine,
and
dl-/3,@-dimethyl-a:-amino-n-butyric
acid. A sample of d( -)-alanine
was
generously provided by Professor M. S. Dunn of the University
of California, Los Angeles.
SUMMARY
1. Extracts of Neurospora contain a d-amino acid oxidase similar in its
action to the d-amino acid oxidase of mammalian tissues.
2. The pH optimum of the system lies at pH 8.0 to 8.5.
3. The enzyme is destroyed by drying, but is not inhibited by cyanide,
iodoacetate, or benzoate.
It is competitively
inhibited by isovaline.
4. The d forms of the following amino acids are rapidly oxidized: methicitrulline,
arginine, a-amino-n-butyric
onine, phenylalanine,
norvaline,
acid, leucine, norleucine, isoleucine, and glutamic acid. The following are
slowly oxidized: aspartic acid, valine, alanine, a-amino-n-caprylic
acid,
lysine, ac-aminophenylacetic
acid, tryptophane,
ornithine,
N-methylleucine. The following are not oxidized : glycine, serine, threonine, proline,
p-ala&e,
,&amino-n-butyric
acid, a-amino-a!-ethylbutyric
acid, ~3,pdimethyl-oc-amino-n-butyric
acid, N , N-dimethylleucine,
leucylglycine, and
isovaline.
5. The activity of the enzyme shows a marked dependence on the chain
length of the substrate.
It was found that an optimum
chain length
exists.
6. The role of d-amino acid oxidase in the wild type and in mutants of
Neurospora is discussed.
BIBLIOGRAPHY
1. Bonner, D., Tatum, E. L., and Beadle, G. W., Arch.
2. Tatum, E. L., Bonner, D., and Beadle, G. W., Arch.
3. Krebs, H. A., Biochem.
J., 29, 1620 (1935).
4. Horowitz, N. H., and Beadle, G. W., J. Biol. Chem.,
5. Waelsch, H., and Borek, E., J. Am. Chem. Sot., 61,
6. Cahill, W. M., and Rudolph, G. G., J. Biol. Chem.,
Biochem.,
Biochem.,
$71
(1943).
3, 477 (1944).
160, 325 (1943).
2252 (1939).
146, 201 (1942).
Downloaded from www.jbc.org at CALIFORNIA INSTITUTE OF TECHNOLOGY on December 26, 2006
Serine is only one-half as effective as I-serine in promoting this reaction in
experiments in viva (15), indicating that Neurospora is unable to convert
d- to I-serine.
This finding is in harmony with the observation
that
Similar evidence for
d-serine is not attacked by the Neurospora enzyme.
other amino acids has been obtained with mutants currently under investigation and will be published at a later date.
N.
HOROWITZ
149
Klein, J. R., and Kamin, H., J. Biol. Chem., 138,507 (1941).
Snyder, F. H., and Corley, R. C., J. Biol. Chem., 122,491 (1937-38).
Rodney, G., and Garner, R. L., J. Biol. Chem., 126, 209 (1938).
Klein, J. R., and Handler, P., J. Biol. Chem., 139, 103 (1941).
Lineweaver, H., and Burk, D., J. Am. Chem. Sot., 66,658 (1934).
Keilin, D., and Hartree, E. F., Proc. Roy. Sot. London, Series B, 119, 114 (1936).
Shemin, D., and Rittenberg, D., J. BioZ. Chem., 161, 507 (1943).
Taturn, E. L., and Bonner, D. M., J. BioZ. Chem., 161,349 (1943).
Tatum, E. L., and Bonner, D., Proc. Nut. Acad. SC., 30, 30 (1944).
Downloaded from www.jbc.org at CALIFORNIA INSTITUTE OF TECHNOLOGY on December 26, 2006
7.
8.
9.
10.
11.
12.
13.
14.
15.
H.